Classification of complex motions from captured sensor data may be a complex and computationally costly procedure. While algorithms may exist for quantifying simple user actions, such as step counting, more complex actions still lack accuracy in quantification. In particular, in the realm of action sports, such as skateboarding, snowboarding, surfing, skiing, and the like, few, if any, algorithms exist for classifying and quantifying maneuvers performed by sports participants using appropriate sports equipment, such as skateboards, snowboards, or the like. With the introduction of these sports into the Olympics and other high level competitions, automatic methods for real- or near-real time maneuver classification and scoring has become increasingly important, for example, for judging competitions in action sports.